Prompt Engineering Professional
Certifies professional mastery of prompt engineering for language models: design, optimisation, API integration and output quality evaluation.
Key skills
- ✓Design and optimisation of prompts for LLMs (GPT, Claude, Mistral)
- ✓Advanced techniques: few-shot, chain-of-thought, RAG, function calling
- ✓Integration of LLM APIs into business workflows
- ✓Evaluation, benchmarking and management of hallucinations
The question bank for this exam is prepared after your purchase. The exam is available no later than 48 hours after payment; you will receive an email as soon as it opens.
Is this right for you?
✓ This certification is right for you if:
- →Software developers and engineers integrating generative AI
- →Data Scientists and ML Engineers specialised in NLP
- →Product Managers and consultants in AI transformation
✗ This certification is not suitable if:
- →Beginners without technical experience: prioritise first a training in programming and machine learning fundamentals.
- →ML experts seeking specialisation in fine-tuning or model training: consult our Master 801 certifications in ML Engineering.
Prerequisites & eligibility
Online registration, secure payment, identity verification during the proctored exam.
Purchase options
Exam voucher, LMS pathway, bundle packs — build your order
Options d'achat
Composez votre parcours — voucher d'examen, préparation LMS, packs entreprise.
Skills assessed
Prompt Design
Mastery of design patterns and optimisation techniques to maximise LLM output quality.
Technical integration
Implementation of OpenAI, Anthropic, Mistral APIs and orchestration with LangChain or Semantic Kernel.
RAG and context
Design of Retrieval-Augmented Generation systems with embeddings and vector databases.
Quality evaluation
Evaluation metrics, hallucination detection and iterative performance improvement.
Exam format
Curriculum
LLM fundamentals and prompt engineering
Transformer architecture and attention mechanisms. Tokenization process (BPE, WordPiece). Inference parameters: temperature, top-p, top-k, frequency penalty. Context management and limits (4K to 200K tokens). Model overview: GPT-4 Turbo, Claude 3 Opus/Sonnet, Mistral Large, Gemini Pro, Llama 3. Proprietary vs open-source differences. Safety and content filtering concepts.
Advanced prompting techniques
Zero-shot prompting: effective formulation of instructions. Few-shot learning: selection and order of examples. Chain-of-thought (CoT): decomposition of reasoning. Self-consistency and voting mechanisms. ReAct pattern: reasoning + acting. Tree-of-thoughts for multi-branch exploration. Prompt chaining and decomposition of complex tasks. Instruction engineering and output format (JSON, XML). Constraint techniques and guardrails.
Integration and orchestration
OpenAI APIs: chat completions, embeddings, moderation. Anthropic Claude API and constitutional AI. Mistral AI and European models. LangChain: chains, agents, memory, callbacks. Microsoft's Semantic Kernel. Function calling and tool use. Streaming and asynchronous management. Token management and cost optimisation. Rate limiting and retry strategies. Monitoring and logging of interactions. Security: prompt injection, data leakage.
RAG and contextual systems
Retrieval-Augmented Generation: principles and architecture. Embeddings: text-embedding-ada-002, voyage-2, multilingual-e5. Vector databases: Pinecone, Weaviate, Qdrant, ChromaDB, FAISS. Chunking strategies: size, overlap, semantic splitting. Metadata and filtering. Hybrid search: dense + sparse retrieval. Reranking with cross-encoders. Query transformation and HyDE. RAG evaluation: relevance, faithfulness, answer correctness. Management of hallucinations and source citation. Multi-query and retrieval fusion.
How to fund your certification
3 ways to cover the cost of your certification
OPCO — Employer funding
Ask your employer to cover the cost through your OPCO (employee training plan).
Personal — €380 incl. VAT
Secure online payment by credit card (Stripe).
Business rate: €500 excl. VAT
Business — custom quote
Teams, cohorts or private catalogue: a quote tailored to your volumes and goals.
Frequently asked questions
Which LLM models are covered by the certification?
The exam covers the main commercial models (GPT-4, Claude 3, Gemini) and open-source models (Mistral, Llama 3). The principles taught are agnostic and applicable to all modern LLMs.
Do I need to know how to code to take this certification?
A basic understanding of code (particularly Python) is strongly recommended for the API integration and RAG parts.
Does this certification cover model fine-tuning?
No, the focus is on prompt engineering and API usage. Fine-tuning and model training are covered by other PROVA certifications in ML Engineering.
Your certification pathway
Before, during, after — the recommended progression
Droits du candidat
Transparence
Les critères de décision et le barème sont documentés et accessibles avant l'examen.
Droit d'appel
Toute décision peut être contestée dans les 30 jours. Examen par un tiers indépendant.
Plainte
Toute personne peut signaler un dysfonctionnement. Formulaire public accessible sans compte.
